Reconciliacion mpios PAR: 2017-2019
Felipe Gonzalez-Arango
22 de julio de 2019
1 Setup
Con este codigo se importan todas las bases de datos y todos los paquetes necesarios.
#ANTES DE CORRER, ¡CORRER LA PESTANA "CODIGO PARA ARRANCAR TODO"!
#Set Working Directory
setwd("C:/Users/felig/Dropbox/Proyecto Juan Camilo")
rm(list=ls())
#Importar base de datos donde esta todo
library(haven)
library(readstata13)
library(tidyr)
library(plyr)
library(dplyr)
library(gridExtra)
library(ggplot2)
library(forcats)
load("C:/Users/felig/Dropbox/Proyecto Juan Camilo/MergeBases_Environment.RData")
base_stata <- read_dta("Para Stata/base_acdi_stata.dta",
encoding="UTF-8")
base_stata_2019 <- read_dta("C:/Users/felig/Dropbox/Proyecto Juan Camilo/Para Stata/base_acdi_stata_2019.dta", encoding="UTF-8")2 Donde la categoria son municipios homicidios
2.1 Separandolos por media
x2017 <- base_stata %>% select(reconciliacion_mean, disculpas_mean, violencia_mean, confianza_vec_mean,confianza_medios_mean, confianza_instituciones_mean)
x2019 <- base_stata_2019 %>% select(ends_with("mean2019"), ends_with("men2019"))
par(mar=c(4,2,0.1,0.1))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_homicidio=(ifelse(homi_cienmil_mean>mean(homi_cienmil_mean),
"Alta homicidios","Baja homicidios"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_homicidio)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
par(mar=c(4,2,0.1,0.1))
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_homicidio=(ifelse(homi_cienmil_mean>mean(homi_cienmil_mean),
"Alta homicidios","Baja homicidios"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_homicidio)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
2.2 Separandolos en Q1 y Q3
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_homicidio=as.factor((ifelse
(homi_cienmil_mean>quantile(homi_cienmil_mean, 0.75),
"Alta homicidios",
ifelse(homi_cienmil_mean<quantile(homi_cienmil_mean, 0.25),
"Baja homicidios",
NA))))) %>%
filter(mun_homicidio!=is.na(mun_homicidio)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_homicidio)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017",y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
par(mfrow=c(2,4))
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_homicidio=as.factor((ifelse
(homi_cienmil_mean>quantile(homi_cienmil_mean, 0.75),
"Alta homicidios",
ifelse(homi_cienmil_mean<quantile(homi_cienmil_mean, 0.25),
"Baja homicidios",
NA))))) %>%
filter(mun_homicidio!=is.na(mun_homicidio)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_homicidio)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
2.3 Separandolos entre el maximo y el minimo
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_homicidio=as.factor((ifelse
(homi_cienmil_mean==max(homi_cienmil_mean),
"Alta homicidios",
ifelse(homi_cienmil_mean==min(homi_cienmil_mean),
"Baja homicidios",
NA))))) %>%
filter(mun_homicidio!=is.na(mun_homicidio)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_homicidio)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_homicidio=as.factor((ifelse
(homi_cienmil_mean==max(homi_cienmil_mean),
"Alta homicidios",
ifelse(homi_cienmil_mean==min(homi_cienmil_mean),
"Baja homicidios",
NA))))) %>%
filter(mun_homicidio!=is.na(mun_homicidio)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_homicidio)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
3 Donde la categoria son municipios violentados
3.1 Separandolos por media
par(mar=c(4,2,0.1,0.1))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_violentado=(ifelse(Ataques_Pobl_Civil_mean>mean(Ataques_Pobl_Civil_mean),
"violentado","No violentado"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
x <- base_stata_2019 %>% select(ends_with("mean2019"), ends_with("men2019"))
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_violentado=(ifelse(Ataques_Pobl_Civil_mean>mean(Ataques_Pobl_Civil_mean),
"violentado","No violentado"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
3.2 Separandolos en Q1 y Q3
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_violentado=as.factor((ifelse
(Ataques_Pobl_Civil_mean>quantile(Ataques_Pobl_Civil_mean, 0.75),
"violentado",
ifelse(Ataques_Pobl_Civil_mean<quantile(Ataques_Pobl_Civil_mean, 0.25),
"No violentado",
NA))))) %>%
filter(mun_violentado!=is.na(mun_violentado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_violentado=as.factor((ifelse
(Ataques_Pobl_Civil_mean>quantile(Ataques_Pobl_Civil_mean, 0.75),
"violentado",
ifelse(Ataques_Pobl_Civil_mean<quantile(Ataques_Pobl_Civil_mean, 0.25),
"No violentado",
NA))))) %>%
filter(mun_violentado!=is.na(mun_violentado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
3.3 Separandolos entre el maximo y el minimo
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_violentado=as.factor((ifelse
(Ataques_Pobl_Civil_mean==max(Ataques_Pobl_Civil_mean),
"violentado",
ifelse(Ataques_Pobl_Civil_mean==min(Ataques_Pobl_Civil_mean),
"No violentado",
NA))))) %>%
filter(mun_violentado!=is.na(mun_violentado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_violentado=as.factor((ifelse
(Ataques_Pobl_Civil_mean==max(Ataques_Pobl_Civil_mean),
"violentado",
ifelse(Ataques_Pobl_Civil_mean==min(Ataques_Pobl_Civil_mean),
"No violentado",
NA))))) %>%
filter(mun_violentado!=is.na(mun_violentado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
4 Donde la categoria son municipios con presencia de victimas (secuestro)
4.1 Separandolos por media
par(mar=c(4,2,0.1,0.1))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_secuestrado=(ifelse(secuestro_cienmil_mean>mean(secuestro_cienmil_mean),
"Alto secuestro","Bajo secuestro"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_secuestrado)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_secuestrado=(ifelse(secuestro_cienmil_mean>mean(secuestro_cienmil_mean),
"Alto secuestro","Bajo secuestro"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_secuestrado)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
4.2 Separandolos en Q1 y Q3
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_secuestrado=as.factor((ifelse
(secuestro_cienmil_mean>quantile(secuestro_cienmil_mean, 0.75),
"Alto secuestro",
ifelse(secuestro_cienmil_mean<quantile(secuestro_cienmil_mean, 0.25),
"Bajo secuestro",
NA))))) %>%
filter(mun_secuestrado!=is.na(mun_secuestrado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_secuestrado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017",y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
par(mfrow=c(2,4))
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_secuestrado=as.factor((ifelse
(secuestro_cienmil_mean>quantile(secuestro_cienmil_mean, 0.75),
"Alto secuestro",
ifelse(secuestro_cienmil_mean<quantile(secuestro_cienmil_mean, 0.25),
"Bajo secuestro",
NA))))) %>%
filter(mun_secuestrado!=is.na(mun_secuestrado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_secuestrado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
4.3 Separandolos entre el maximo y el minimo
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_violentado=as.factor((ifelse
(secuestro_cienmil_mean==max(secuestro_cienmil_mean),
"Alto secuestro",
ifelse(secuestro_cienmil_mean==min(secuestro_cienmil_mean),
"Bajo secuestro",
NA))))) %>%
filter(mun_violentado!=is.na(mun_violentado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017",y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_violentado=as.factor((ifelse
(secuestro_cienmil_mean==max(secuestro_cienmil_mean),
"Alto secuestro",
ifelse(secuestro_cienmil_mean==min(secuestro_cienmil_mean),
"Bajo secuestro",
NA))))) %>%
filter(mun_violentado!=is.na(mun_violentado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019",y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
5 Donde la categoria son municipios con presencia de desmoviliazdos
5.1 Separandolos por media
par(mar=c(4,2,0.1,0.1))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_desmovilizado=(ifelse(desmovilizados_cienmil_mean>mean(desmovilizados_cienmil_mean),
"Alto desmovilizado","Bajo desmovilizado"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_desmovilizado)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_desmovilizado=(ifelse(desmovilizados_cienmil_mean>mean(desmovilizados_cienmil_mean),
"Alto desmovilizado","Bajo desmovilizado"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_desmovilizado)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
5.2 Separandolos en Q1 y Q3
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_desmovilizado=as.factor((ifelse
(desmovilizados_cienmil_mean>quantile(desmovilizados_cienmil_mean, 0.75),
"Alto desmovilizado",
ifelse(desmovilizados_cienmil_mean<quantile(desmovilizados_cienmil_mean, 0.25),
"Bajo desmovilizado",
NA))))) %>%
filter(mun_desmovilizado!=is.na(mun_desmovilizado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_desmovilizado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017",y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_desmovilizado=as.factor((ifelse
(desmovilizados_cienmil_mean>quantile(desmovilizados_cienmil_mean, 0.75),
"Alto desmovilizado",
ifelse(desmovilizados_cienmil_mean<quantile(desmovilizados_cienmil_mean, 0.25),
"Bajo desmovilizado",
NA))))) %>%
filter(mun_desmovilizado!=is.na(mun_desmovilizado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_desmovilizado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019",y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
5.3 Separandolos entre el maximo y el minimo
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_violentado=as.factor((ifelse
(desmovilizados_cienmil_mean==max(desmovilizados_cienmil_mean),
"Alto desmovilizado",
ifelse(desmovilizados_cienmil_mean==min(desmovilizados_cienmil_mean),
"Bajo desmovilizado",
NA))))) %>%
filter(mun_violentado!=is.na(mun_violentado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017",y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_violentado=as.factor((ifelse
(desmovilizados_cienmil_mean==max(desmovilizados_cienmil_mean),
"Alto desmovilizado",
ifelse(desmovilizados_cienmil_mean==min(desmovilizados_cienmil_mean),
"Bajo desmovilizado",
NA))))) %>%
filter(mun_violentado!=is.na(mun_violentado)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_violentado)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019",y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
6 Donde la categoria son municipios pobres
6.1 Separandolos por media
par(mar=c(4,2,0.1,0.1))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_pobre=(ifelse(nbi_mean>mean(nbi_mean),
"Alta pobreza","Baja pobreza"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_pobre)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_pobre=(ifelse(nbi_mean>mean(nbi_mean),
"Alta pobreza","Baja pobreza"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_pobre)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
6.2 Separandolos en Q1 y Q3
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_pobre=as.factor((ifelse
(nbi_mean>quantile(nbi_mean, 0.75),
"Alta pobreza",
ifelse(nbi_mean<quantile(nbi_mean, 0.25),
"Baja pobreza",
NA))))) %>%
filter(mun_pobre!=is.na(mun_pobre)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_pobre)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017",y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_pobre=as.factor((ifelse
(nbi_mean>quantile(nbi_mean, 0.75),
"Alta pobreza",
ifelse(nbi_mean<quantile(nbi_mean, 0.25),
"Baja pobreza",
NA))))) %>%
filter(mun_pobre!=is.na(mun_pobre)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_pobre)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019",y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
6.3 Separandolos entre el maximo y el minimo
par(mfrow=c(2,4))
x <- base_stata[,165:207] %>% select(ends_with("mean"))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_pobre=as.factor((ifelse
(nbi_mean==max(nbi_mean),
"Alta pobreza",
ifelse(nbi_mean==min(nbi_mean),
"Baja pobreza",
NA))))) %>%
filter(mun_pobre!=is.na(mun_pobre)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_pobre)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017",y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_pobre=as.factor((ifelse
(nbi_mean==max(nbi_mean),
"Alta pobreza",
ifelse(nbi_mean==min(nbi_mean),
"Baja pobreza",
NA))))) %>%
filter(mun_pobre!=is.na(mun_pobre)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_pobre)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019",y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
7 Donde la categoria son municipios con presencia de establecimientos educativos
7.1 Separandolos por media
par(mar=c(4,2,0.1,0.1))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_establ=(ifelse(t_establ_mean>mean(t_establ_mean),
"Alta presencia Establecimientos Educativos","Baja presencia Establecimientos Educativos"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_establ)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_establ=(ifelse(t_establ_mean>mean(t_establ_mean),
"Alta presencia Establecimientos Educativos","Baja presencia Establecimientos Educativos"))) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_establ)) +
geom_bar(stat="identity", fill = "white")+
theme_minimal()+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
7.2 Separandolos en Q1 y Q3
par(mfrow=c(2,4))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_establ=as.factor((ifelse
(t_establ_mean>quantile(t_establ_mean, 0.75),
"Alta presencia Establecimientos Educativos",
ifelse(t_establ_mean<quantile(t_establ_mean, 0.25),
"Baja presencia Establecimientos Educativos",
NA))))) %>%
filter(mun_establ!=is.na(mun_establ)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_establ)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017", y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_establ=as.factor((ifelse
(t_establ_mean>quantile(t_establ_mean, 0.75),
"Alta presencia Establecimientos Educativos",
ifelse(t_establ_mean<quantile(t_establ_mean, 0.25),
"Baja presencia Establecimientos Educativos",
NA))))) %>%
filter(mun_establ!=is.na(mun_establ)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_establ)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019", y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
7.3 Separandolos entre el maximo y el minimo
par(mfrow=c(2,4))
x <- base_stata[,165:207] %>% select(ends_with("mean"))
for(i in 1:length(x2017)){
plot <-
base_stata %>%
filter(dummyPAR==1) %>%
mutate(mun_establ=as.factor((ifelse
(t_establ_mean==max(t_establ_mean),
"Alta presencia Establecimientos Educativos",
ifelse(t_establ_mean==min(t_establ_mean),
"Baja presencia Establecimientos Educativos",
NA))))) %>%
filter(mun_establ!=is.na(mun_establ)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2017)[i]]], col=mun_establ)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2017",y=names(x2017[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
} 
for(i in 1:length(x2019)){
plot <-
base_stata_2019 %>%
mutate(mun_establ=as.factor((ifelse
(t_establ_mean==max(t_establ_mean),
"Alta presencia Establecimientos Educativos",
ifelse(t_establ_mean==min(t_establ_mean),
"Baja presencia Establecimientos Educativos",
NA))))) %>%
filter(mun_establ!=is.na(mun_establ)) %>%
ggplot(aes(x=Municipio, y=.[[names(x2019)[i]]], col=mun_establ)) +
geom_bar(stat="identity", na.rm=TRUE, fill="white")+
scale_fill_manual(values = c("#00AFBB", "#E7B800"))+
labs(title="Mpios PAR para el 2019",y=names(x2019[i]))+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
} 
8 Pruebas T
base_stata_2017 <- base_stata %>% filter(dummyPAR==1)
for (i in 1:length(x2017)){
ttest <- t.test(as.data.frame(base_stata_2017[names(x2017)][i]),as.data.frame(base_stata_2019[names(x2019)][i]))
print(ttest)
}##
## Welch Two Sample t-test
##
## data: as.data.frame(base_stata_2017[names(x2017)][i]) and as.data.frame(base_stata_2019[names(x2019)][i])
## t = 6.794, df = 40.548, p-value = 3.401e-08
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 2.155696 3.980260
## sample estimates:
## mean of x mean of y
## 3.3107970 0.2428193
##
##
## Welch Two Sample t-test
##
## data: as.data.frame(base_stata_2017[names(x2017)][i]) and as.data.frame(base_stata_2019[names(x2019)][i])
## t = 14.187, df = 51.939, p-value < 2.2e-16
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.8496715 1.1296346
## sample estimates:
## mean of x mean of y
## 0.8295794 -0.1600736
##
##
## Welch Two Sample t-test
##
## data: as.data.frame(base_stata_2017[names(x2017)][i]) and as.data.frame(base_stata_2019[names(x2019)][i])
## t = -1.3726, df = 51.917, p-value = 0.1758
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.4086848 0.0766831
## sample estimates:
## mean of x mean of y
## -1.256594 -1.090593
##
##
## Welch Two Sample t-test
##
## data: as.data.frame(base_stata_2017[names(x2017)][i]) and as.data.frame(base_stata_2019[names(x2019)][i])
## t = 5.2054, df = 37.421, p-value = 7.255e-06
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.2185224 0.4968931
## sample estimates:
## mean of x mean of y
## 2.505096 2.147388
##
##
## Welch Two Sample t-test
##
## data: as.data.frame(base_stata_2017[names(x2017)][i]) and as.data.frame(base_stata_2019[names(x2019)][i])
## t = 7.8051, df = 34.305, p-value = 4.134e-09
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 3.572252 6.086258
## sample estimates:
## mean of x mean of y
## 18.36326 13.53401
##
##
## Welch Two Sample t-test
##
## data: as.data.frame(base_stata_2017[names(x2017)][i]) and as.data.frame(base_stata_2019[names(x2019)][i])
## t = 12.234, df = 33.506, p-value = 6.573e-14
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 11.76371 16.45352
## sample estimates:
## mean of x mean of y
## 39.50485 25.396239 Graficos generales
#Creacion de la bsae de datos para poder comparar los graficos por ano.
ano2017 <- c()
ano2017[1:27] <- 2017
ano2017 <- t(ano2017)
ano2019 <- c()
ano2019[1:27] <- 2019
ano2019 <- t(ano2019)
#Aqui juntamos los valores de interes de la ola 2017 y 2019.
base_stata_2019_tidy <- cbind(base_stata_2019[1],base_stata_2017[names(x2017)], base_stata_2019[names(x2019)])
#Aqui se hace el tidy para poder generar la variable "indicador"
base_stata_2019_tidy_ensayo <- base_stata_2019_tidy %>% gather(key=indicador, value=valor, -Municipio)## Warning: attributes are not identical across measure variables;
## they will be dropped#Creamos la columna para distinguir las olas
base_stata_2019_tidy_ensayo$ano <- 0
base_stata_2019_tidy_ensayo[1:162,4] <- 2017
base_stata_2019_tidy_ensayo[163:324,4] <- 2019
#Limpiamos nombres para que todos encajen en uno mismo.
base_stata_2019_tidy_ensayo$indicador <- gsub("_mean","", base_stata_2019_tidy_ensayo$indicador, ignore.case = TRUE)
base_stata_2019_tidy_ensayo$indicador <- gsub("2019","", base_stata_2019_tidy_ensayo$indicador, ignore.case = TRUE)
base_stata_2019_tidy_ensayo$indicador <- gsub("_men","", base_stata_2019_tidy_ensayo$indicador, ignore.case = TRUE)
base_stata_2019_tidy_ensayo %>%
ggplot(aes(x=as.factor(indicador), y=valor, col=as.factor(ano))) +
geom_point()+
facet_wrap(~Municipio)+
labs(title="Comparacion Mpios PAR 2017 y 2019")+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
base_stata_2019_tidy_ensayo %>%
ggplot(aes(x=Municipio, y=valor, col=as.factor(ano))) +
geom_point()+
facet_wrap(~as.factor(indicador))+
labs(title="Comparacion Mpios PAR 2017 y 2019")+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
10 Graficos por municipio y por indicador (se hace zoom)
#Volver categoricas las variables categoricas.
base_stata_2019_tidy_ensayo$indicador <- as.factor(base_stata_2019_tidy_ensayo$indicador)
for (i in 1:length(unique(base_stata_2019_tidy_ensayo$Municipio))){
plot <- base_stata_2019_tidy_ensayo %>%
ggplot(aes(x=as.factor(ano), y=valor, col=Municipio[i])) +
geom_point()+
facet_wrap(~indicador)+
labs(title="Comparacion Mpios PAR 2017 y 2019")+
theme(plot.title = element_text(hjust = 0.5), axis.text.x = element_text(angle=90))
print(plot)
}
